By: Mark Glennon*
The Food and Drug Administration last week published results from the first batch of COVID-19 antibody diagnostics to have their accuracy independently evaluated by federal laboratories. They looked at the 12 blood tests it already authorized for emergency use.
Tests made by Abbott Laboratories and Roche scored particularly well and are highly likely to give accurate results, according to the evaluation. As summarized by FierceBiotech, “Abbott’s test “showed a sensitivity of 100%, specificity of 99.6%, and positive and negative predictive values of 92.9% and 100%, respectively.” Roche’s test “showed 100% sensitivity and 99.8% specificity across multiple antibody types, with a positive predictive value of 96.5% and a negative predictive value of 100% at 5% prevalence.”
Antibody tests, also called serology tests, are important because they indicate how much of the population had coronavirus, though they probably never knew, and are therefore likely to be immune for some period of time. A surprisingly large prevalence of antibodies would indicate that the effective fatality is smaller than thought.
Though some other states had already been using antibody testing to inform policy, Governor JB Pritzker on April 24 announced that Illinois would not be running antibody tests and he has released no results from tests done so far. We criticized that in our April 27 article.
Why isn’t that being reported more broadly? Why isn’t that being discussed in Illinois’ daily briefings on the virus?
Legitimate concerns about antibody tests stem from some low-quality tests on the market with high rates of inaccuracy, leading to the danger of individuals unwisely relying on positive results to think they are immune. Additionally, it’s not completely certain whether, or for how long, antibodies provide immunity.
However, those concerns never were pertinent to the research and policy value of antibody testing. Researchers can correct for known error levels, according to those we cited in our earlier article. Results of the testing in some other states have shown a surprisingly large portion of the population with antibodies.
Surprisingly large prevalence of antibodies in turn has a huge effect on projections by making the number of likely future infections smaller, shifting the policy balance in favor of loosened restrictions.
The high accuracy rate of Abbott Labs’ test should come as no surprise. It has been claiming 99% accuracy since early April. Abbott is not the kind of company that would take the liability and reputational risk of such a claim if it did not have solid evidence. The FDA now confirmed that it does.
In short, Illinois never did have a good excuse for not collecting and publishing antibody data. Now it has none. On April 24 Pritzker said, “As soon as [antibody tests] prove themselves accurate and reliable, I will make it a priority to get them into our communities as widely as we can.” The press should hold him to that and asking about those Elgin and Roseland results, but don’t hold your breath.
That note takes us to a different topic: Will Pritzker, the press and Dr. Ngozi Ezike (Illinois’s director of public health) please get their story straight about unimportant but seemingly alarming “new case” numbers released daily?
Pritzker last week finally started saying what we’ve long said, which is that daily new case numbers mean little. That’s because, as Pritzker said, new cases are “really a function of how many tests we do.” Testing has exploded, now reaching as much as 17,000 per day in Illinois. More tests mean more reported cases.
But what’s the first thing Dr. Ezike announces every day? New case numbers. Her department’s daily press releases are headlined like this one from May 6: Public Health Officials Announce 2,270 New Cases of Coronavirus Disease.
The press parrots that misleadingly scary message back with similar headlines every day. Google “new cases Illinois” and see four yourself. On May 4 NBC Chicago ran a story about Pritzker explaining why new case numbers don’t matter much. But what was their headline a few hours later? “Illinois Reports 2,341 New Coronavirus Cases, Lifting State Total to 63,840.” Do they read their own articles?
We repeat our standard warning that none of this is meant to imply that the Wuhan virus isn’t dangerous. It is. But we think that threat is confined almost entirely to identifiable, at-risk groups in particularly areas and that a better balance must be struck against the economic catastrophe being inflicted by the shutdowns. We agree with the Wall Street Journal’s Friday editorial:
The tradeoff isn’t between lives and livelihoods. The policy goal has to be to protect both as much as possible. Deploy more personal protective equipment, greatly increase testing, build surge capability to handle flare-ups, and isolate society’s most vulnerable to keep hospitals from getting overwhelmed. But for heaven’s sake reopen the economy so we don’t consign millions to years of poverty.
UPDATE: This article has been updated to delete reference to a Chicago Tribune report on antibody testing at Greater Elgin Family Care Center. Our reference incorrectly said they found a 60% positive testing rate for antibodies, which in fact was the rate for the test for the infection itself.
*Mark Glennon is founder of Wirepoints.